COPYRIGHT NOTICE

This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.

Adaptive Signal ProcessingMaster levelNo. of credits: 7

Textbooks: Sayed, A.H.,Fundamentals of Adaptive Filtering, Wiley, 2003

Haykin, S.,Adaptive Filter Theory, 4/E, Prentice-Hall, 2001

J.C. Principe et al.,Neural and Adaptive Systems,Wiley, 2000Tutorials:Linear adaptive filtering ( Chapter 3 from my DSP book, in Romanian)

Introduction to Artificial Neural Networks (Chapter 1 from my ANN book, in Romanian)General descriptionThe course focuses on advanced topics on adaptive filtering. Main themes are related to linear filtering algorithms in time and frequency domains, nonlinear adaptive filters implemented as neural networks, neural architectures and learning algorithms for feedforward and recurrent networks. Applications include pattern recognition (OCR and face processing), data transmission channel equalization and analog decoding, biomedical time series analysis, system identification. Software support is provided by MATLAB and NeuroSolutions neural networks simulator.

Course outline General introduction to adaptive linear filtering

Optimal filtering problem and Wiener solution. Definition and characterization of random processes

Classification criteria of adaptive algorithms. Cost functions.

Applications of adaptive filtersFirst order adaptive algorithms: gradient descent, Newton algorithm. LMS algorithm and its variants.

Adaptive filtering in the frequency domainLeast-Squares (LS) algorithm. Recurrent Least-Squares (RLS) algorithm. Lecture 5: General introduction to Artificial Neural Networks (ANN's)

Motivations for studying ANN's

Definition and classification criteria for ANN's

Applications of ANN'sLecture 6: Multilayer perceptron (MLP)

Standard backpropagation training algorithm

Variants of backpropagation algorithmLecture 7: Radial Basis Functions (RBF) networkLecture 8-9: Recurrent neural networks

Analog systems: Hopfield network - theory and applications, Cellular Neural Networks (CNN)

Discrete systems: Hopfield and Elman networks - theory and applicationsLecture 10-12: Applications of ANN's

Pattern Recognition

Channel equalization

Digital filter design